System and method for near real-time analysis of interests and sentiment
AssigneeColossio, Inc.
InventorJoseph A. Jaroch
Life of this patent
Estimating the term…
Abstract
A system for analyzing user sentiment includes one or more processors, one or more memory components, one or more network connections, and first and second databases. Further according to this aspect, the first database includes a plurality of concepts partially developed by processing a plurality of data repositories, and the second database includes a user record partially developed by processing data related to post content. Still further, the one or more processors search the second database by one or more interests to determine one or more user interest levels, and the one or more interests are correlated with one or more of the plurality of concepts.
Claims20
A system for analyzing user sentiment, comprising:
- one or more processors;
- one or more memory components;
- one or more network connections;
- and
- first and second databases;
- wherein the first database includes a plurality of concepts partially developed by processing a plurality of data repositories;
- wherein the second database includes a user record partially developed by processing data related to post content;
- wherein the one or more processors search the second database by one or more interests to determine one or more user interest levels;
- and
- wherein the one or more interests are correlated with one or more of the plurality of concepts.
A system for tracking sentiment in near real-time, comprising:
- a processor;
- a memory component;
- and
- wherein the system is operatively coupled to a network;
- wherein a plurality of interests are stored on the memory component;
- wherein the plurality of interests are arranged according to a plurality of database identification numbers;
- wherein the system receives user content from a plurality of social media platforms;
- wherein the system applies to the user content an organizational scheme based on the plurality of interests and the plurality of database identification numbers;
- one or more user records developed by the processor and stored on the memory component;
- wherein the one or more user records represent chronologically organized sentiments of one or more users across the plurality of social media platforms.
A method for analyzing user sentiment, comprising:
- receiving data from a plurality of web sources;
- performing natural language processing on the plurality of web sources;
- identifying n-grams corresponding to a plurality of interests within the web sources;
- assigning one or more database identification numbers to the plurality of interests;
- receiving user content;
- organizing the user content chronologically according to a time period when the user content was generated by a user;
- analyzing the user content to identify n-grams corresponding to the plurality of interests;
- assigning to the user content the one or more database identification numbers depending on the n-grams identified within the user content;
- tracking a sentiment score of the user content when the one or more database identification numbers are identified within the user content;
- developing one or more user records wherein the one or more database identification numbers, the sentiment scores, and timestamps correlating to receipt of the user content are saved together.
Dependencies are inferred from the claim text.